The response of salt marsh vegetation to tidal reduction caused by the Oosterschelde storm-surge barrier

In 1986 a sluice gate barrier was completed in the mouth of the Oosterschelde estuary. The barrier has been partially or completely closed during 1986 and the first months of 1987. Consequently the high tides were reduced to such a level that the salt marshes were scarcely flooded. Since April 1987 the barrier has been closed on average during two out of 706 high tides a year. Although the barrier allows tidal exchange the tidal flow has been restricted as a result of the reduced width of the mouth of the estuary. This restriction of the tidal flow caused a 26 cm decrease of mean high water in the eastern part of the estuary. As a result the inundation frequencies of the salt marshes decreased.

The response of salt-marsh vegetation to this tidal reduction was analyzed using annual records (1982 till 1990) of species composition in 57 permanent plots in two marshes at the southern shore of the estuary. Analysis of the response of individual species to marsh elevation in the pre and the post-barrier situation revealed that most species moved down the marsh elevation gradient. The first axis of an ordination (DCA-1) was significantly negatively related to inundation frequency. Between 1984 and 1990 all plots were displaced towards a significantly higher ordination score, indicating a trend towards a species composition from higher up the marsh. The position of most plots along DCA-1 remained stable until 1985 and started to increase in 1986 or 1987. The vegetation in plots dominated either by Halimione portulacoides Footnote1 or by Spartina anglica, started to change in 1985. This premature change was attributed to frost damage in January 1985. The initially high rate of change along DCA-1 decreased in 1989 and 1990. This would suggest that the vegetation re-equilibrated with the newly established tidal conditions. Further analysis revealed no significant difference in the relation between inundation frequency and sample score along DCA-1 between 1984 and 1990. This corroborates the view that species composition had re-equilibrated with the tidal conditions. Along DCA-2 the samples were displaced towards a significantly higher score as well. The change was attributed to an increase of the perennial Halimione and the annual Suaeda maritima. The annual Atriplex hastata displayed an increase from 1986 till 1988, but strongly declined in 1990. The transient response of this annual was described by DCA-3.

During the pre-barrier phase several attempts had been made to predict the vegetation response to tidal reduction. We developed a multivariate model (CCA) with inundation frequency as a constraining factor to describe the relation between inundation frequency and species composition in the pre-barrier phase in 1984. Next we used this model to predict the score of the samples along CCA-1 in the post-barrier phase from the inundation frequencies in 1990. The actual ordination score in 1990 was calculated from the observed species composition. The predicted ordination scores did not differ significantly from the observed ones. In retrospect it is concluded that the response of the vegetation to the reduction of the tides, as far as the fraction of species composition which is related to inundation frequency is concerned, could have been predicted by our model.